alternative interlingua

R. V. Guha <>
Date: Fri, 17 Aug 90 10:15 CDT
From: R. V. Guha <>
Subject: alternative interlingua
In-Reply-To: The message of 17 Aug 90 09:13 CDT from Peter F. Patel-Schneider <>
Message-id: <19900817151514.1.GUHA@GAIA.ACA.MCC.COM>
    Date: Fri, 17 Aug 90 10:13:04 EDT
    From: (Peter F. Patel-Schneider)

    > Posted-Date: Mon, 13 Aug 90 10:39 CDT
    > Date: Mon, 13 Aug 90 10:39 CDT
    > From: R. V. Guha <ai.guha@MCC.COM>
    >     Date: Mon, 13 Aug 90 11:36:28 EDT
    >     From: (Peter F. Patel-Schneider)
    >     I tried to pick a number of features that I thought were useful for KR, and
    >     that had a chance of being formalized.  The idea is that the logic would be
    >     extended as necessary.
    >     The way to translate axioms with abs in them would be to turn them into
    >     default rules, not to leave the abs predicates in.
    > I don't understand how default rules came into the discussion. I was 
    > using the abs + a circumscriptive axiom to minimize the abs.
    > Guha

    What are the abs things for?  They certainly are representationally
    suspect.  I always think of them as logical "connectives" for defaults of
    some sort.


You are telling me how I should represent something and I don't think the interlingua
should try to do such things.

abs are real relations (as real as father or mother) for me and they certainly
should not be lost in the process of translation.